MINISTRY OF ECONOMY Guidelines for the Regulation Impact Assessment
62
Annex 4 – Ideas about the questionnaire
Below are some tips on how to make the consultation-supporting questionnaire.
• Aim and scope of consultation should be presented in a concise way.
• Ask for identification data to sort out the respondents into appropriate groups and
include the personal data protection clause.
• Add instruction wherever necessary.
• Questions should be ordered from general to specific.
• Leave some space for notes and comments at the bottom of the questionnaire.
• Always voice your thanks for filling out the questionnaire form.
• At the end of the questionnaire, ask the respondent for the permission to contact him
again to get more information.
• Make the questions as short as possible.
• Avoid double negation.
• It is good to ask a control question, that is, a rephrased question asked in some other
place of the questionnaire. Different answers may mean that the respondent did not
understand the problem asked about. Do not include both questions in the analysis.
• A question should ask about only one subject, e.g., “Have you tried to get information
from public administration but you were not successful in getting comprehensive info?
Yes/No.” If the answer is “no,” you will never know whether the respondent did not
look for information at all, or he did but did not get any.
• Ask precise questions: avoid hazy phrases, such as, good, sufficient. Instead of asking:
“Was access to information sufficient?” you can better ask: “Was information
available on the Internet, TV, newspapers?”
• Questions about the past should define the time period. Instead of asking: “how
often?” you can better ask “how many times in a month/year?”
• If not all the questions should be asked of all the respondents, insert filter questions,
e.g., “If you have not used welfare programmes, go to question 5.”
• Avoid questions suggesting the right answer, e.g., “You do not agree with….?”
• Avoid closed questions. They encourage short answers but the alternative of “yes” or
“no” only may lead you to wrong conclusions.
• It is good to obtain answers presenting various degrees of intensity of any
phenomenon. Example: “I totally agree/I agree/Undecided/I do not agree/I totally do
not agree.” You may also use a scale, e.g., from 1 to 10.
• Avoid graphics as much as you can as this may slow down uploading the
questionnaire from the Internet.
• Use the right colours and font type.
• Leave enough space for the answers, especially to open questions, so as the
respondent can freely express himself/herself.
Do'stlaringiz bilan baham: |